######################################
The Python Performance Benchmark Suite
######################################
The ``pyperformance`` project is intended to be an authoritative source of
benchmarks for all Python implementations. The focus is on real-world
benchmarks, rather than synthetic benchmarks, using whole applications when
possible.
* `pyperformance documentation `_
* `pyperformance GitHub project `_
(source code, issues)
* `Download pyperformance on PyPI `_
pyperformance is distributed under the MIT license.
Documentation:
.. toctree::
:maxdepth: 2
usage
benchmarks
custom_benchmarks
cpython_results_2017
changelog
Other Python Benchmarks:
* CPython: `speed.python.org `_ uses pyperf,
pyperformance and `Codespeed `_ (Django
web application)
* PyPy: `speed.pypy.org `_
uses `PyPy benchmarks `_
* Pyston: `pyston-perf `_
and `speed.pyston.org `_
* `Numba benchmarks `_
* Cython: `Cython Demos/benchmarks
`_
* pythran: `numpy-benchmarks
`_
See also the `Python speed mailing list
`_ and the `Python pyperf module
`_ (used by pyperformance).
pyperformance is not tuned for PyPy yet: use the `PyPy benchmarks project
`_ instead to measure PyPy
performances.
Image generated by bm_raytrace (pure Python raytrace):
.. image:: images/bm_raytrace.jpg
:alt: Pure Python raytracer